Description of Project 2 Hemophilia, an inherited bleeding disorder, is currently treated by protein replacement therapy that provides only transient correction and is a significant financial burden due to the need for frequent infusions. For those with access to concentrate, the most serious complication in the management of the disorder is the development of neutralizing antibodies to the infused factor that occur in up to 30% of those with hemophilia A. For such patients, control of bleeds by infusion of high-dose recombinant activated Factor Vll (rFVlla) results in effective liemostasis. However, this treatment carries a substantial cost due to the short half-life of the protein and the need for repeated dosing. To address these shortcomings we previously developed a novel gene transfer approach whereby we can effect hemostasis using a FVII transgene, engineered to be secreted in the active, two chain form (FVIIa). Using the liver as the target tissue, we demonstrated that continuous expression of FVlla following adeno-associated virus (AAV) gene transfer can correct the phenotypic defect in hemophilic mice and dogs. This approach has immunological advantages over Factor IX (FIX) or Factor Vlll (FVlll) gene strategies since the recipient is tolerant to the transgene product. Moreover it can effect hemostasis even in the face of neutralizing antibodies to FVIII or FIX. Additionally, since rFVlla is used for hemostasis management for other coagulation defects (such a platelet disorders), in contrast to FVIII/FIX gene-based approaches, FVIIa gene transfer may unify the treatment of several coagulation defects to a single product. However, a substantial proportion of hemophilia patients have advanced liver disease due to viral hepatitis, thus rendering the liver an unsuitable tissue for viral vector-mediated gene transfer. We developed a method for transducing extensive areas of muscle by an AAV vector expressing biologically active canine FIX and demonstrated multi- year hemostatic efficacy in hemophilia B dogs. Therefore, as a theme for this project, we propose to use skeletal muscle as the target tissue for expression and secretion of FVlla, using the canine model(s) of hemophilia that closely mimic the human condition. Initially, we wish to perform a systematic study of promoters and AAV serotypes for muscle expression of FVlla, and study the hemostatic efficacy and safety of canine FVlla (cFVlla) expression under optimal conditions, including dogs with inhibitors (Aim 1). In subsequent experiments, we will determine the therapeutic index of a variant of cFVlla with increased activity, as a means to reduce the effective vector dose/viral load per cell and thus address any potential limitations of muscle to perform post-translational modifications necessary for cFVlla biological activity, due to increased antigen expression (Aim 2). Lastly, we will develop a regulated system for cFVlla expression from muscle via a small, orally bioavailable drug (doxycycline) that will enhance the safety of FVIIa gene transfer (Aim 3). Overall, the experiments proposed in the project will establish in a comprehensive manner the pre-clinical basis for muscle-derived FVlla gene transfer.